Project Abstract Chemotherapy has been documented to be efficacious in prolonging survival for patients with various malignancies. One critical finding from numerous clinical trials and pooled analyses is that the efficacy of chemotherapy for breast cancer has been shown to decrease with age, from a 27% proportional reduction in 10-year mortality for women less than 50 years of age to 14% for women ages 50-59 years and 8% for women ages 60-69 years, and no benefit for women 70 years or older. These findings have remained constant even when the pooled results were examined with the addition of more patients from more recent trials in 1998, 2001, 2005 and 2008. These results are striking because there has been no decreasing efficacy of chemotherapy observed with advancing age for women with ovarian cancer, and for men and women with lung cancer and colon cancer. Numerous clinical trials have shown that chemotherapy is equally efficacious across different age groups of men and women with lung and colon cancer as well as women with ovarian cancer, including populations 70 years or older. Therefore, evidence-based clinical guidelines on chemotherapy from the National Institutes of Health (NIH) and other health authorities recommend chemotherapy for women with breast cancer less than 70 years old, whereas there are no age-restricted recommendations for patients with colon, ovarian and lung cancer. In other words, chemotherapy is strongly recommended for patients of all age groups with colon, ovarian and lung cancer. Randomized controlled clinical trials are considered to be the gold standard for determining the efficacy of a therapy. However, participants in clinical trials often do not represent a cross-section of patients in the community, and older patients were significantly under-represented in clinical trials. Population-based observational studies can therefore be useful adjuncts to randomized controlled trials in determining whether efficacy under controlled conditions in specialty centers translates into real-world effectiveness in the community and provide valuable insight into the benefit of a therapy in community practices that cannot be obtained from the clinical trials.Furthermore, in some circumstances where the randomized trials are not possible due to ethical or logistical issues, well-conducted observational studies can potentially serve to determine the comparative effectiveness of various drug therapies. Therefore, there is a critical need for information on whether the effectiveness of chemotherapy decreases with age for breast cancer only (as shown in clinical trials), or also decreases with age for lung, colorectal, and ovarian cancer and on whether chemotherapy is cost effective by different age groups and by tumor site. The specific aims of this proposal are to: 1) determine whether the stage-specific efficacy of chemotherapy observed in controlled clinical trials is translated into real-world effectiveness in prolonging survival among community dwelling patients with breast, ovarian, lung and colorectal cancer at age 65 years or older; 2) determine the association between the effectiveness of chemotherapy and advancing age (i.e., to determine whether chemotherapy is equally effective for all age groups of patients with these tumors); 3) determine the comparative effectiveness of various chemotherapy agents or combination regimen-specific benefits in prolonging survival and their associations with age and tumor type; 4) determine whether and to what extent patient and tumor characteristics as well as hospital and physician factors affect the relationship between the effectiveness of chemotherapy and age; and 5) to determine the cost- effectiveness of chemotherapy by age and tumor type. Our approach is to identify a large nationwide, population-based cohort of over 550,000 patients diagnosed with breast, ovarian, lung and colorectal cancer at age 65 years or older in 1991 through 2005 (with more cancer cases in 2006- 2007 from the new data linkage in early 2011) from the 17 Surveillance, Epidemiology, and End Results (SEER) registries, accounting for over 25% of the U.S. population. The large numbers of patients will enable us to examine the comparative effectiveness of chemotherapy in many subpopulations. This proposed study will examine the comparative- and cost-effectiveness of chemotherapy in association with advancing age across four major tumors in the real world community setting, which has never been tested before. The proposed innovative use of the nationwide and population-based Medicare claims data, which provide information on specific chemotherapy drug or regimens, will also allow for chemotherapy-specific effectiveness analysis.